The Lomax distribution also known as Pareto distribution of the second kind or Pearson Type VI distribution has been used in the analysis of income data, and business failure data. It may describe the lifetime of a decreasing failure rate component as a heavy tailed alternative to the exponential distribution. In this paper we consider the estimation of the parameter of Lomax distribution. Baye’s estimator is obtained by using Jeffery’s and extension of Jeffery’s prior by using squared error loss function, Al-Bayyati’s loss function and Precautionary loss function. Maximum likelihood estimation is also discussed. These methods are compared by using mean square error through simulation study with varying sample sizes. The study aims to find out a suitable estimator of the parameter of the distribution. Finally, we analyze one data set for illustration.

1. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
IJSM
Bayesian analysis of shape parameter of Lomax
distribution using different loss functions
1
Afaq Ahmad, 1
S.P Ahmad and 2
A. Ahmed
1
Department of Statistics, University of Kashmir, Srinagar, India
2
Department of Statistics and Operation Research, Aligarh Muslim University, India
Corresponding Author: Afaq Ahmad, Department of Statistics, University of Kashmir, Srinagar, India, Tel:
+919596169720, Email: baderaafaq@gmail.com
The Lomax distribution also known as Pareto distribution of the second kind or Pearson Type VI
distribution has been used in the analysis of income data, and business failure data. It may
describe the lifetime of a decreasing failure rate component as a heavy tailed alternative to the
exponential distribution. In this paper we consider the estimation of the parameter of Lomax
distribution. Baye’s estimator is obtained by using Jeffery’s and extension of Jeffery’s prior by
using squared error loss function, Al-Bayyati’s loss function and Precautionary loss function.
Maximum likelihood estimation is also discussed. These methods are compared by using mean
square error through simulation study with varying sample sizes. The study aims to find out a
suitable estimator of the parameter of the distribution. Finally, we analyze one data set for
illustration.
Keywords: Lomax distribution, Bayesian estimation, priors, loss functions, fisher information matrix.
INTRODUCTION
The Lomax distribution also known as Pareto distribution of second kind has, in recent years, assumed opposition of
importance in the field of life testing because of its uses to fit business failure data. It has been used in the analysis of
income data, and business failure data. It may describe the lifetime of a decreasing failure rate component as a heavy
tailed alternative to the exponential distribution. Lomax distribution was introduced by Lomax (1954), Abdullah and
Abdullah (2010), estimates the parameters of Lomax distribution based on Generalized probability weighted moment.
Zangan (1999) deals with the properties of the Lomax distribution with three parameters. Abd-Elfatth and Mandouh
(2004) discussed inference for R = Pr{Y<X} when X and Y are two independent Lomax random variables. Nasiri and
Hosseini (2012) performs comparisons of maximum likelihood estimation (MLE) based on records and a proper prior
distribution to attain a Bayes estimation (both informative and non-informative) based on records under quadratic loss
and squared error loss functions. The cumulative distribution function of Lomax distribution is given by
)1(11),:(











x
xF
Therefore, the corresponding probability density function is given by
International Journal of Statistics and Mathematics
Vol. 2(1), pp. 055-065, January, 2015. © www.premierpublishers.org ISSN: 2375-0499 x
Research Article

2. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Ahmad et al. 055
)2(0,,0;1),;(
)1(













x
x
xf
where  and  are shape and scale parameters, respectively.
The survival function is given by
)3(1),;(











x
xR
and the hazard function is given by
)4(1),;(
)12( 











x
xh
This paper is arranged as follows: Section 2, 3 and 4 discusses the Bayesian methodology using Jeffrey’s prior and
extension of Jeffrey’s prior information under different loss functions for estimation of the shape parameter of Lomax
distribution with known scale. Section 5, focuses in the simulation study and results to compare the estimators and
finally section 6 is the conclusion of the paper.
MATERIAL AND METHODS
Prior and Loss Functions
Recently Bayesian estimation approach has received great attention by most researchers. Bayesian analysis is an
important approach to statistics, which formally seeks use of prior information and Bayes Theorem provides the formal
basis for using this information. In this approach, parameters are treated as random variables and data is treated fixed.
An important pre-requisite in Bayesian estimation is the appropriate choice of prior(s) for the parameters. However,
Bayesian analysts have pointed out that there is no clear cut way from which one can conclude that one prior is better
than the other. Very often, priors are chosen according to ones subjective knowledge and beliefs. However, if one has
adequate information about the parameter(s) one should use informative prior(s); otherwise it is preferable to use non
informative prior(s). In this paper we consider the extended Jeffrey’s prior proposed by Al-Kutubi (2005) as:
     
 RcIg
c
 1,1
where     








 2
2
;log



xf
nEI is the Fisher’s information matrix. For the model (2),
 
1
2
c
n
kg 






where k is a constant, with the above prior, we use three different loss functions for the model (2), first is the squared
error loss function which is symmetric, second is Albayyati,s loss function and third is the precautionary loss function
which is a simple asymmetric loss function.
It is well known that choice of loss function is an integral part of Bayesian inference. As there is no specific analytical
procedure that allows us to identify the appropriate loss function to be used, most of the works on point estimation and
point prediction assume the underlying loss function to be squared error which is symmetric in nature. However, in-
discriminate use of SELF is not appropriate particularly in these cases, where the losses are not symmetric. Thus in
order to make the statistical inferences more practical and applicable, we often needs to choose an asymmetric loss
function. A number of asymmetric loss functions have been shown to be functional, see Zellner (1986), Chandra (2001)
etc. In the present work, we consider symmetric as well as asymmetric loss functions for better comprehension of
Bayesian analysis.
a) The first is the common squared error loss function given by:

3. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Int. J. Stat. Math. 056
    )5(ˆ,ˆ 2
1   cl
which is symmetric,  and ˆ represent the true and estimated values of the parameter. This loss function is frequently
used because of its analytical tractability in Bayesian analysis.
b) The second is the Al-Bayyati’s loss function of the form
)6(,)(),( 2
22
Rcl c



c) The third is the precautionary loss function given by:
    )7(
ˆ
ˆ
,ˆ
2




l
which is an asymmetric loss function, for details, see Norstrom (2012). This loss function is interesting in the sense that
a slight modification of squared error loss introduces asymmetry.
Maximum Likelihood Estimation
In this section we consider maximum likelihood estimators (MLE) of Lomax distribution. Let x1, x2,…, xn be a random
sample of size n from Lomax distribution, then the log likelihood function can be written as








n
i
x
nnL
1
1ln)1(lnln),(ln


As the parameter  is assumed to be known, the ML estimator of  is obtained by solving the equation
0
),(ln




L
01ln
1






 
n
i
xn

)8(
1ln
1









 n
i
ML
x
n


Bayesian estimation of Lomax distribution under Jeffrey’s prior by using different loss function:
Consider n recorded values, ),...,,( 21 nxxxx  having probability density function as
01),;(
)1(








x
x
xf




The likelihood function is given by















n
i
n
x
xL
1
)1(
1)/(




Thus, in our problem we consider the prior distribution of  to be
)()(  Ig 

4. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Ahmad et al. 057
where  k-vector is valued parameter and )(I is the Fisher’s information matrix of order .kk  For the model (2) the
prior distribution is given by


1
)( g
The posterior distribution of  is given by
)()/()/(  gxLx 















n
i
n
x
x
1
)1(
1
1)/(
















 
 n
i
n
n
x
x
1
1
1ln)1(exp)/(

















 

n
i
n x
kx
1
1
1lnexp)/(


where k is independent of 
and 

 d
x
k
n
i
n
 
















0 1
11
1lnexp
n
n
i
x
n
k

















1
1
1ln

n
x
k
n
n
i















1
1ln

Hence posterior distribution of  is given by



























 



n
i
n
n
n
i x
n
x
x
1
11
1lnexp
1ln
)/(




  )9(exp)/( 1
t
n
t
x n
n
 

 
where 







n
i
x
t
1
1ln

Estimator under squared error loss function
By using squared error loss function
2
1 )(),(  

cl for some constant c1 the risk function is given by






),()(  lER
 dxl
 

0
)/(),(

5. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Int. J. Stat. Math. 058
   dt
n
t
c n
n


 
 
 exp)( 12
0
1
      







 







 dtdtdt
n
tc nnn
n
exp2expexp
0
11
0
12
0
11
2





 





 



12
1 )1(
2
)2(
2
nnn
n
t
n
t
n
t
n
n
tc

t
nc
t
nnc
c






 1
2
1
2
1
2)1(
Now solving 0
)(






R
, we obtain the Baye’s estimator as
0
2)1( 1
2
1
1
2

















t
nc
t
nnc
c



0
2
2 1
1 

t
nc
c 
t
n
s 


)10(
1ln
1









 n
i
s
x
n


Estimation under Al-Bayyati’s Loss Function
Al-Bayyati, [5] introduced a new loss function given as
2
)(),( 2
 

c
l . Employ this loss function, we obtain the
Baye’s estimator under Jeffery prior information.
By using this loss function, we obtained the following risk function
 dxR c
)/()()( 2
0
2



   dt
n
t
R n
n
c


 

 exp)()( 12
0
2







0
12
)exp()()( 2
 dt
n
t
R cn
n





 






t
cn
t
cn
cn
nt
R c
)1(
2
)2(
)(
1
)( 2
2
2
2
2
2


6. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Ahmad et al. 059
Now solving 0
)(






R
, we obtain the Baye’s estimator as
0
)1(
2
)2(
)(
1 2
2
2
2
2
2












 




 

t
cn
t
cn
cn
ntc


0
)1(2
)(2
1 2
22





 




t
cn
cn
ntc

t
cn
A
2



)11(
1ln
1
2










 n
i
A
x
cn


Estimation under Precautionary Loss Function:
For determining the Baye’s estimator of  we will introduce asymmetric precautionary loss function given by


 




2
)(
),(l
By using this loss function the risk function is given by
 









0
1
2
exp
)(
)( 


 dt
n
t
R n
n
 








0
122
exp)()( 

 dt
n
t
R n
n
t
nc
t
nnc
c 1
2
1
1
2)1(


 



Now solving 0
)(






R
, we obtain the Baye’s estimator as
  )12(
1ln
)1(
1
2
1










 n
i
p
x
nn


Bayesian estimation of Lomax distribution under the extension of Jeffrey’s prior by using different loss
function
We consider the extended Jeffery prior is taken as
  
 RcIg
c
,)()( 

7. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Int. J. Stat. Math. 060
where   







 2
),;(log
)(



xf
nEI is the Fisher’s information matrix. For the model (2) the prior distribution of  is
given by
c
g 2
1
)(

 
The posterior distribution of  is given by















n
i
c
n
x
x
1
2
)1(
1
1)/(
















 
 n
i
n
cn
x
x
1
2
1ln)1(exp)/(

















 

n
i
cn x
kx
1
2
1lnexp)/(


where k is independent of 
and 

 d
x
k
n
i
cn
 
















0 1
21
1lnexp
12
1
1
1ln
)12(



















cn
n
i
x
cn
k

)12(
1ln
12
1


















cn
x
k
cn
n
i 
Hence posterior distribution of  is given by
  )13(exp
)12(
)/( 2
12
t
cn
t
x cn
cn
 

 

where 

oftindependenis
x
t
n
i








1
1ln
Estimator under squared error loss function
By using squared error loss function
2
1 )(),(  

cL for some constant c the risk function is given by
   dt
cn
t
cR cn
cn


 
 
 exp
)12(
)()( 2
12
2
0
1



























 





n
i
cn
cn
n
i x
cn
x
x
1
2
12
1
1lnexp
)12(
1ln
)/(





8. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Ahmad et al. 061





 





 



123212
12
1 )12(
2
)32()12(
)12(
2
cncncn
cn
t
cn
t
cn
t
cn
cn
tc

t
cnc
t
cncnc
c
)12(2)12)(22( 1
2
12
1








Now solving 0
)(






R
, we obtain the Baye’s estimator as
)14(
1ln
1212
1












 n
i
ss
x
cn
t
cn


Remark 1.1: Replacing c= 1/2 in (14), we get the same Bayes estimator as obtained in (10) corresponding to the
Jeffrey’s prior.
Estimation under Al-Bayyati’s Loss Function
By using the Al-Bayyati’s loss function the risk function is given by
 dt
cn
t
R cn
cn
c
)exp(
)12(
)()( 2
12
0
22


 
 








0
22
12
)exp()(
)12(
)( 2
 dt
cn
t
R ccn
cn





 





 



22
2
32
2
12
22
12
222
)22(
2
)32()12(
)12(
)( ccnccnccn
cn
t
ccn
t
ccn
t
ccn
cn
t
R 





 






t
ccn
t
ccn
ccn
cntc
)22(
2
)32(
)12(
)12(
1 2
2
2
2
2
2

Now solving 0
)(






R
, we obtain the Baye’s estimator as
0
)22(
2
)32(
)12(
)12(
1 2
2
2
2
2
2











 




 

t
ccn
t
ccn
ccn
cntc


)15(
1ln
1212
1
22












 n
i
AA
x
ccn
t
ccn


Remark 1.2: Replacing c= 1/2 in (15), we get the same Bayes estimator as obtained in (11) corresponding to the
Jeffrey’s prior.
Estimation under Precautionary loss function
By using precautionary loss function 

 




2
)(
),(l the risk function is given by

9. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Int. J. Stat. Math. 062
  


 dt
cn
t
R cn
cn



 




 exp
)12(
)(
)( 2
12
0
2
t
cn
t
cncn
R
)12(2)12)(22(
)(
2



 



Now solving 0
)(






R
, we obtain the Baye’s estimator as
  )16(
1ln
)12)(22(
1
2
1










 n
i
p
x
cncn


Remark 1.3: Replacing c= 1/2 in (16), we get the same Bayes estimator as obtained in (12) corresponding to the
Jeffrey’s prior.
Simulation Study
In our simulation study, we chose a sample size of n=25, 50 and 100 to represent small, medium and large data set. The
shape parameter is estimated for Lomax distribution with Maximum Likelihood and Bayesian using Jeffrey’s & extension
of Jeffrey’s prior methods. For the shape parameter we have considered  = 0.5 and 1.5. The values of Jeffrey’s
extension were c = 0.5 and 1.0. The value for the loss parameter c2 =  1.0 and 2.0. This was iterated 5000 times and
the shape parameter for each method was calculated. A simulation study was conducted R-software to examine and
compare the performance of the estimates for different sample sizes with different values for the Extension of Jeffreys’
prior and the loss functions. The results are presented in tables for different selections of the parameters and c
extension of Jeffrey’s prior.
Table 1. Mean Squared Error for )(

 under Jeffery’s prior
n  
ML



S A


c2=1
A


c2= -1
A


c2= 2.0
A


c2 = -2.0
p


25 0.5 1.0 0.0118 0.0113 0.0129 0.0106 0.0154 0.0108 0.0120
1.5 1.0 0.1298 0.1255 0.1492 0.1106 0.1815 0.1043 0.1362
0..5 2.0 0.0111 0.0107 0.0090 0.0131 0.0079 0.0161 0.0098
1.5 2.0 0.0872 0.0840 0.0777 0.0966 0.0780 0.1156 0.0801
50 0.5 0.5 0.0050 0.0049 0.0050 0.0049 0.0053 0.0052 0.0049
1.5 1.5 0.0811 0.0800 0.0918 0.0704 0.1059 0.0630 0.0856
0.5 0.5 0.0056 0.0055 0.0059 0.0052 0.0066 0.0052 0.0057
1.5 1.5 0.0680 0.0682 0.0591 0.0786 0.0514 0.0904 0.0635
100 0.5 0.5 0.0024 0.0024 0.0024 0.0024 0.0024 0.0025 0.0024
1.5 1.5 0.0752 0.0751 0.0691 0.0813 0.0634 0.0880 0.0721
0.5 0.5 0.0028 0.0028 0.0030 0.0027 0.0033 0.0026 0.0029
1.5 1.5 0.0242 0.0240 0.0225 0.0259 0.0214 0.0282 0.0232
ML= Maximum Likelihood, S=SELF, A= Al-Byatti’s Loss Function, P= Precautionary Loss Function
In Table 1, Bayes estimation with Al-Bayytai’s Loss function under Jeffrey’s prior provides the smallest values in
most cases especially when loss parameter C2 is -1, - 2.

10. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Ahmad et al. 063
Table 2. Mean Squared Error for )(

 under extension of Jeffery’s prior
n   C
ML



S A


c2=1
A


c2=-1
A


c2=2
A


c2=-2
p


25 0.5 1.0 0.5 0.0118 0.0113 0.0129 0.0106 0.0154 0.0108 0.0120
0.5 1.0 1.0 0.0143 0.0117 0.0133 0.0110 0.0160 0.0112 0.0124
1.5 1.0 0.5 0.1007 0.0979 0.0822 0.1192 0.0721 0.1461 0.0894
1.5 1.0 1.0 0.0872 0.0934 0.0808 0.1124 0.0745 0.1379 0.0863
0.5 2.0 0.5 0.0260 0.0102 0.0087 0.0124 0.0079 0.0152 0.0094
0.5 2.0 1.0 0.0158 0.0134 0.0157 0.0120 0.0191 0.0117 0.0144
1.5 2.0 0.5 0.1073 0.1045 0.0866 0.1281 0.0739 0.1568 0.0949
1.5 2.0 1.0 0.0977 0.0880 0.0900 0.0939 0.0994 0.1074 0.0881
50 0.5 1.0 0.5 0.0052 0.0050 0.0054 0.0050 0.0059 0.0051 0.0052
0.5 1.0 1.0 0.0059 0.0067 0.0058 0.0078 0.0051 0.0090 0.0063
1.5 1.0 0.5 0.2795 0.2781 0.3147 0.2443 0.0354 0.2137 0.2959
1.5 1.0 1.0 0.0972 0.1094 0.0960 0.1240 0.0833 0.1399 0.1026
0.5 2.0 0.5 0.0112 0.0112 0.0098 0.0127 0.0085 0.1445 0.0105
0.5 2.0 1.0 0.0077 0.0087 0.0075 0.0103 0.0065 0.0115 0.0081
1.5 2.0 0.5 0.1225 0.1212 0.1398 0.1051 0.1608 0.0914 0.1301
1.5 2.0 1.0 0.1525 0.1682 0.1514 0.1859 0.1356 0.2048 0.1597
100 0.5 1.0 0.5 0.0025 0.0025 0.0025 0.0025 0.0027 0.0025 0.0025
0.5 1.0 1.0 0.0024 0.0025 0.0023 0.0026 0.0023 0.0028 0.0024
1.5 1.0 0.5 0.0229 0.0226 0.0216 0.0240 0.0211 0.0258 0.0221
1.5 1.0 1.0 0.0516 0.0564 0.0513 0.0618 0.0466 0.0675 0.0538
0.5 2.0 0.5 0.0027 0.0026 0.0027 0.0025 0.0029 0.0025 0.0026
0.5 2.0 1.0 0.0025 0.0026 0.0025 0.0028 0.0023 0.0031 0.0025
1.5 2.0 0.5 0.0736 0.7377 0.0808 0.0664 0.0890 0.0601 0.0770
1.5 2.0 1.0 0.0225 0.0222 0.0221 0.0228 0.0223 0.0239 0.0221
ML= Maximum Likelihood, S=SELF, A= Al-Bayytai’s Loss Function, P= Precautionary Loss Function
In Table 2, Bayes estimation with Al-Bayytai’s Loss function under extension of Jeffrey’s prior provides the smallest
values in most cases especially when loss parameter C2 is -1,-2 whether the extension of Jeffrey’s prior is 0.5, 1.0.
Moreover, when the sample size increases from 25 to 100, the MSE decreases quite significantly.
Application
Here, we use a real data set to compare the estimates of the Lomax distribution. We consider an uncensored data set
corresponding an uncensored data set from consisting of 100 observations on breaking stress of carbon fibers (in Gba):
3.7, 2.74, 2.73, 2.5, 3.6, 3.11, 3.27,2.87, 1.47, 3.11,4.42, 2.41, 3.19, 3.22, 1.69, 3.28, 3.09, 1.87, 3.15, 4.9, 3.75, 2.43,
2.95, 2.97, 3.39, 2.96, 2.53,2.67, 2.93, 3.22, 3.39, 2.81, 4.2, 3.33, 2.55, 3.31, 3.31, 2.85, 2.56, 3.56, 3.15, 2.35, 2.55,
2.59,2.38, 2.81, 2.77, 2.17, 2.83, 1.92, 1.41, 3.68, 2.97, 1.36, 0.98, 2.76, 4.91, 3.68, 1.84, 1.59, 3.19,1.57, 0.81, 5.56,
1.73, 1.59, 2, 1.22, 1.12, 1.71, 2.17, 1.17, 5.08, 2.48, 1.18,3.51, 2.17, 1.69,1.25, 4.38, 1.84, 0.39, 3.68, 2.48, 0.85, 1.61,
2.79, 4.7, 2.03, 1.8, 1.57, 1.08, 2.03, 1.61, 2.12,1.89, 2.88, 2.82, 2.05, 3.65. These data are used here only for
illustrative purposes. The required numerical evaluations are carried out using the Package of R software.
We used the non-informative Jeffery’s priors for θ when  = 1, 2, values of Jeffrey’s extension were c = 1.0 and when c
= 0.5 it gives same estimates as in case of Jeffery’s prior (Table 3). The value of loss parameters c=(1, -1, 2, -2) The
estimated results are given at Table 3 and 4.

11. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Int. J. Stat. Math. 064
Table 3: Estimates incase of Jefffery’s prior

ML



S A


c2=1
A


c2= -1
A


c2= 2.0
A


c2 = -2.0
p


1.0 0.0978 0.0977 0.1026 0.0929 0.1076 0.0882 0.1001
2.0 0.5471 0.5469 0.5650 0.5291 0.5834 0.5116 0.5560
Table 4. Estimates in case of extension of Jeffery’s prior
 C
ML



S A


c2=1
A


c2=-1
A


c2=2
A


c2= -2
p


1.0 1.0 0.0978 0.0929 0.0977 0.0882 0.1026 0.0837 0.0953
2.0 1.0 0.5471 0.5290 0.5468 0.5115 0.5649 0.4943 0.5379
From these tables we conclude that Al-Bayytai’s loss is best among these as it gives the smallest values of estimates
especially when c2 = -1 , -2.
CONCLUSION
In this article, we have primarily studied the Bayes estimator of the parameter of the Lomax distribution under the
Jeffery’s and extended Jeffrey’s prior assuming three different loss functions. The extended Jeffrey’s prior gives the
opportunity of covering wide spectrum of priors to get Bayes estimates of the parameter - particular cases of which are
Jeffrey’s prior and Hartigan’s prior. We have also addressed the problem of Bayesian estimation for the Lomax
distribution, under symmetric loss functions and that of Maximum Likelihood Estimation. From the results, we observe
that in most cases, Bayesian Estimator under New Loss function (Al-Bayyati’s Loss function) has the smallest Mean
Squared Error values for both prior’s i.e, Jeffrey’s and an extension of Jeffrey’s prior information.
ACKNOWLEDGMENT
The authors would like to thank the referee for a very careful reading of the manuscript and making a number of nice
suggestions which improved the earlier version of the manuscript.
REFERENCES
Abdullah, MA, Abdullah, HA (2010). Estimation of Lomax parameters based on Generalized Probability Weighted
Moment, JKAU, Sci (22), 171-184.
Abd-Elfattah, M, Mandouh, RM (2004). Estimation of Pr [Y< X] in Lomax case, The 13th Annual Conference in Statistics
Computer Science and Operation Research, ISSR, Cairo University, Egypt, Part, 1, 156-166.
Al-Bayyati (2002). Comparing methods of estimating Weibull failure models using simulation, Ph.D. Thesis, College of
Administration and Economics, Baghdad University, Iraq.
Al-Kutubi, HS (2005). On Comparison Estimation Procedures for parameter and survival function. Iraqi Journal of
Statistical Science, vol. 9, 1-14.
Chandra, M. J., (2001). “Statistical Quality Control”, CRC Press, Boca Raton.
Jeffery’s, H (1964). An invariant form for the Prior Probability in estimation problems, Proceedings of the Royal Society
of London, Series. A., 186: 453-461.
Lomax, HS, (1954). Business Failures, Another example of the analysis of Failure data, JASA, (49), 847-852.

12. Bayesian analysis of shape parameter of Lomax distribution using different loss functions
Ahmad et al. 065
Nasiri, P, Hosseini, S (2012). Statistical Inferences for Lomax Distribution Based on Record Values (Bayesian and
Classical), Journal of Modern Applied Statistical Methods, 11, 179-189.
Norstrom, JG (1996). The use of Precautionary Loss Functions in Risk Analysis, IEEE Transactions on reliability, (3),
400-403.
Reshi, JA, Ahmed, A, Ahmad, SP (2014). Bayesian Analysis of Scale Parameter of the Generalized Inverse Rayleigh
Model Using Different Loss Functions, International Journal of Modern Mathematics, 10(2), pp. 151-162.
Zellner, A, (1986). Bayesian estimation and prediction using asymmetric loss function, J.Amer. Statist. Assoc., 81, pp.
446-451.exponential distribution using simulation. PhD Thesis, Baghdad University, Iraq.
Zagan, AA, (1999). Lomax distribution and its role in reliability and life testing, M.Sc Thesis, King Abdul –Alaziz
University, Saudia Arabia.
Accepted 12 January, 2015.
Citation: Afaq A, SP Ahmad, Ahmed A (2015). Bayesian analysis of shape parameter of Lomax distribution using
different loss functions. Int. J. Stat. Math., 2(1): 055-065.
Copyright: © 2015 Ahmad et al. This is an open-access article distributed under the terms of the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original
author and source are cited.